and now for something completely different part 3

[JoaMat 1,195]

A small step towards a balanced Raal UberAmp setup. 

Completed a balanced smd (mostly) driver board today. Size 99mm x 35mm (3.9in x 1.4in).

 Two through hole unbalanced driver boards for comparison.

[JoaMat 1,195]

Report from cuisine de joamat.

One side of the balanced board works alright – able to set output current and offset is within a millivolt. The other side – output current controllable but servo doesn’t work as desired, offset, varies round plus two and a half volt.

Here is an aerial picture of the output board.

I’m not so happy to dive in with test pins in there… and servo parts on the bottom side. 

…maybe I should be satisfied with one working side out of two possible. 

[starcat 3]

On 2/13/2018 at 7:49 PM, jose said:

My amp module is ready. Now I'm working on a second Psu for the amp (dual-mono) and at the same time I work on the unbal/bal preamp (I have the PSU ready). I want to do a three module system.

Hey Jose, very nice amp. Mind asking which cases are you using here? Is the sink estate enough to cool all the transistors for CFA3 balanced with 4 channels? 

 

Can someone please point me to the latest gerbers for the CFA3 balanced amp - looks like here in this post is the latest revision? Appreciated, thanks!

As for the PSU the GRLV is it, right. Some building it with 24 or 30 volts? 

Here in this post BOMs for the CFA2 and GRLV 30V.

Edited Tuesday at 11:37 AM by starcat

[jose 1,637]

I usually use Modu boxes. They are expensive (I live outside the continent) but they do the job.

That heatsink is overkill for a single CFA. Even for 2 CFA's they are still a bit oversized (BIAS 190mA).

24vDC power supply. Use OPA445.

[starcat 3]

I suspect using here the 1/2w Xicon 273 50PPM resistors shouldn't be a problem instead of the 1/4w Vishay 100PPM from the BOM? I have a couple of the Xicon's laying around... 

Edited Tuesday at 08:13 PM by starcat

[starcat 3]

6 hours ago, jose said:

I usually use Modu boxes. They are expensive (I live outside the continent) but they do the job.

Very nice indeed. I can however find the Modu Slim line only with profile sides but not with sinks. Do you have a link or model number for the one on the right (with the sinks and milled aluminium front)? 

Edited Tuesday at 04:13 PM by starcat

[jose 1,637]

I don't want to be rude or impertinent but please, use your browser and spend some time looking for the information.

MODU

[Pars 1,602]

Is the one on the right a 2U? Looks to be 1U from the pic.

The SlimLine have 10mm sides which would be sufficient for heatsinking for many applications.

[starcat 3]

1 hour ago, jose said:

I don't want to be rude or impertinent but please, use your browser and spend some time looking for the information.

As you can see I have spent more than some time and even posted the link above. It is simply that I do not find a 1U enclosure with sinks on the sides. Might be that it is discontinued or whatever. If you don't want just don't bother answering. Doesn't mean I have not researched the whole site already. 

1 hour ago, Pars said:

Is the one on the right a 2U? Looks to be 1U from the pic.

The SlimLine have 10mm sides which would be sufficient for heatsinking for many applications.

Exactly, the left is the SlimLine with the profile sides and 10mm milled front as in the shop. The right however looks identical in hight but with sinks and a sinked version isn't available in the shop. Might be that he had it modified though with the sinks. Doesn't really look to me like being 2U. 

Edited Tuesday at 08:09 PM by starcat

[jose 1,637]

both are 2u (8cm)

pesante Disspante

Edited Tuesday at 08:51 PM by jose

[JoaMat 1,195]

My guess is both cases are 2U… optical visual illusion makes them 1U? 

 

[jose 1,637]

yes... but is U2. You can see the three modules here

 

 

 

 

 

[starcat 3]

So you are using the R-core transformer for the preamp? Very nice indeed!

[jose 1,637]

Thanks. Although the GRLV is very clean and stable, for preamps I prefer to use R-cores and easier to place in a single chassis since they do not electromagnetically interfere (or very little) with anything. It´s optional so you can perfectly use a toroid

Now that I've seen the preview, it's been a little over a year since Dr. Gilmore sent me a design for a new buffer based on the CFA architecture. At the moment I have not been able to put it into operation. If confinement and anti-virus measures allow it this year I could have it ready.

Edited Wednesday at 07:59 AM by jose
orthography

[Craig Sawyers 3,386]

The key difference between a toroid and an R-core is primary/secondary capacitance and leakage inductance.

R-core has primary and secondary on opposite legs. So the capacitance is low, but the leakage inductance is high.

With a toroid, the capacitance is high, but the leakage inductance is low.

So an R-core (like a split bobbin EI) is better for not coupling RFI between windings. But the leakage inductance means enough stray magnetic field to cause potential hum problems.

A toroid by contrast, because of the tight coupling between the windings, tends to couple RF. But because of the low leakage inductance has a very low external magnetic field.

Basically, horses for courses. I've used both types in various bits of my kit. But since I always use an outboard supply (for non-power amp use) either type has worked just fine.

[Pars 1,602]

Does anyone have a pair of CFP3 boards they want to get rid of?

Also, I see two sets of Gerbers: cfp3rs and cfp3rss. Is the difference the revised servo/current mirror? I'm assuming the cfp3rss is the most recent?

Thanks!

[Pars 1,602]

@kevin gilmoreDo the current gerbers (cfp3rs and cfp3rss) have the servo/current mirror correction done to the CFP2 on them? Or don't they need it? Thanks!

 

[kevin gilmore 1,456]

which ever has the later date. definitely has the latest servo

[ti5002000 2]

6 hours ago, Pars said:

Does anyone have a pair of CFP3 boards they want to get rid of?

Also, I see two sets of Gerbers: cfp3rs and cfp3rss. Is the difference the revised servo/current mirror? I'm assuming the cfp3rss is the most recent?

Thanks!

Hi @Pars

 

I think I have some spare boards with the latest version, that has the servo completely out of the audio path, as Kevin said. Tomorrow I let you know. 

Best 

Tiago 

Edited Wednesday at 11:47 PM by ti5002000
Edit

[Pars 1,602]

Unclear. Going back in the thread, I had posted this, as the schematics I have don't appear to have this fix in place. Shortly after this post, you released new gerbers including all fixes. However, there was an error with a 10K resistor connecting to a device emitter instead of the base. Not sure whether the fixes mentioned were in only referring to this or also included the servo fix. I don't see a pair of 15K resistors, as were used in the CFA2 fix, on the gerbers.

 

[ti5002000 2]

11 hours ago, Pars said:

Unclear. Going back in the thread, I had posted this, as the schematics I have don't appear to have this fix in place. Shortly after this post, you released new gerbers including all fixes. However, there was an error with a 10K resistor connecting to a device emitter instead of the base. Not sure whether the fixes mentioned were in only referring to this or also included the servo fix. I don't see a pair of 15K resistors, as were used in the CFA2 fix, on the gerbers.

 

Sorry @Pars i wasn´t thinking right..What i have is the CFA2...

My mistake..

[kevin gilmore 1,456]

i went back and looked at everything.  there is no servo in the amp part itself because its in the front end.

the latest file which should be posted somewhere above is cfp3rssfixedsssw also attached here.

cfp3rssfixedsssw - CADCAM.ZIP

[Pars 1,602]

Cool. Thanks Kevin!

[MLA 62]

Had a batch of boards made on the above posted file a month or so ago and won't be using them all. You're most welcome to a pair if you want.

[Pars 1,602]

Most kind of you. Yes, I would like a pair. PM me and we can discuss.